This study will compare the risks and complications of the UVC and the PCVC. It is unclear which of these two catheters is safer for babies. This study hypothesizes that the long-erm use of UVC will present fewer infection and other risks than long-term use of PCVC. This incidence of blood stream infection will be compared between grouips of low birthweight infants under 1250 grams at birth randominzed to either long term (8-28 days) umbilical catherization or short term (<7 days) umbilical catheterization followed by percutaneous central venous catheter placement. The primary outcome will be Kaplan-Meier estimates of infection-free line survival. Secondary outcomes will include the incidence of toerh catheter complications, feeding complications and need for additional venous access.